ABSTRACT The worldwide scale-up of antiretroviral therapy (ART) has decreased HIV mortality and improved clinical outcomes in resource-limited settings (RLS), however, 20-30% of patients on ART typically experience detectable viremia after 12 months [1, 2]. As incomplete suppression is known to be a risk factor for developing drug resistance mutations (DRMs) [3], the development of resistance may erode global gains in the provision of ART [4-7]. In most RLS, a public health approach to choosing first-line (1L) and second-line (2L) ART is utilized [8]. For patients who fail a 1L regimen containing a specific NRTI backbone, the 2L NRTI backbone is expected to have preserved activity. However, in settings where detection of failure is delayed, the accumulation of DRMs may result in compromise of the 2L NRTI backbone. In addition to the concerns regarding NRTI susceptibility for 2L regimens, questions remain regarding the optimal time to switch a patient from 1L to 2L ART; for patients who are failing 1L, but have suboptimal adherence, it has been shown that with additional adherence interventions, patients re-suppress and actually had few or no resistance mutations [14]. There is a need to better understand the threshold and consequences of DRMs resulting from different patterns of non-adherence. Finally, little is known about the impact of accumulated DRMs on subsequent 2L outcomes. For this study, we propose to evaluate the DRMs among patients failing 1L ART, examine the patterns of adherence associated with the development of DRMs, and assess the association between DRMs and subsequent response to 2L ART. Through the analysis of DRM we hope to gain insight that can inform 1L regimen recommendations and increase our understanding of the risks associated with sub-optimal adherence on subsequent 2L outcomes. The proposed study will utilize data and samples from patients that have received ART at the Nigerian Institute of Medical Research (NIMR), Jos University Teaching Hospital (JUTH), and University College Hospital in Ibadan (UCH) in Nigeria. Through PEPFAR funding, all 3 treatment centers have been providing comprehensive HIV care and treatment services for over 8 years 21,138 patients currently on ART. Harvard has established the capacity for DRM genotyping at all three of these centers and will provide technical assistance and support.